Related papers: Light Cones For Vision: Simple Causal Priors For V…
In this article we first correct a recent misconception about a topology that was suggested by Zeeman as a possible alternative to his Fine topology. This misconception appeared while trying to establish the causality in the ambient…
Learning visual features from unlabeled image data is an important yet challenging task, which is often achieved by training a model on some annotation-free information. We consider spatial contexts, for which we solve so-called jigsaw…
Detecting poorly textured objects and estimating their 3D pose reliably is still a very challenging problem. We introduce a simple but powerful approach to computing descriptors for object views that efficiently capture both the object…
In this paper, we propose a deep hierarchical attention context model for lossless attribute compression of point clouds, leveraging a multi-resolution spatial structure and residual learning. A simple and effective Level of Detail (LoD)…
This paper concentrates on the homogeneous (conformal) model of Euclidean space (Horosphere) with subspaces that intuitively correspond to Euclidean geometric objects in three dimensions. Mathematical details of the construction and…
Spontaneous self-organization is ubiquitous in systems far from thermodynamic equilibrium. While organized structures that emerge dominate transport properties, universal representations that identify and describe these key objects remain…
Protein-ligand binding prediction is central to virtual screening and affinity ranking, two fundamental tasks in drug discovery. While recent retrieval-based methods embed ligands and protein pockets into Euclidean space for…
Linear topological spaces with partial ordering (linear kinematics) are studied. They are defined by a set of 8 axioms implying that topology, linear structure and ordering are compatible with each other. Most of the results are valid for…
We introduce an analogue of the theory of length spaces into the setting of Lorentzian geometry and causality theory. The r\^ole of the metric is taken over by the time separation function, in terms of which all basic notions are…
People easily recognize new visual categories that are new combinations of known components. This compositional generalization capacity is critical for learning in real-world domains like vision and language because the long tail of new…
Predictive world models that simulate future observations under explicit camera control are fundamental to interactive AI. Despite rapid advances, current systems lack spatial persistence: they fail to maintain stable scene structures over…
Hierarchical text classification (HTC) is a special sub-task of multi-label classification (MLC) whose taxonomy is constructed as a tree and each sample is assigned with at least one path in the tree. Latest HTC models contain three…
Detecting novel objects without class information is not trivial, as it is difficult to generalize from a small training set. This is an interesting problem for underwater robotics, as modeling marine objects is inherently more difficult in…
Few-shot object detection (FSOD) aims to classify and detect few images of novel categories. Existing meta-learning methods insufficiently exploit features between support and query images owing to structural limitations. We propose a…
Adversarial attacks aim to perturb images such that a predictor outputs incorrect results. Due to the limited research in structured attacks, imposing consistency checks on natural multi-object scenes is a promising yet practical defense…
The appearance of the same object may vary in different scene images due to perspectives and occlusions between objects. Humans can easily identify the same object, even if occlusions exist, by completing the occluded parts based on its…
Humans perceive the 3D world as a set of distinct objects that are characterized by various low-level (geometry, reflectance) and high-level (connectivity, adjacency, symmetry) properties. Recent methods based on convolutional neural…
World models compress rich sensory streams into compact latent codes that anticipate future observations. We let separate agents acquire such models from distinct viewpoints of the same environment without any parameter sharing or…
Modern computer vision applications rely on learning-based perception modules parameterized with neural networks for tasks like object detection. These modules frequently have low expected error overall but high error on atypical groups of…
Most of cosmological observables are light-propagated. I will present coordinates adapted to the propagation of null-like signals as observed by a geodesic observer. These "geodesic light-cone (GLC) coordinates" are general, adapted to…